Comparison of two discrimination indexes in the categorisation of continuous predictors in time-to-event studies

Identificador:  RP:2455

Handle: https://hdl.handle.net/20.500.11797/RP2455

Autors:  Arostegui, Inmaculada; Esteban, Cristóbal; Meira-Machado, Luis; Rodríguez-Álvarez, María Xosé; Barrio, Irantzu

Resum:
The Cox proportional hazards model is the most widely used survival prediction model for analysing time-to-event data. To measure the discrimination ability of a survival model the concordance probability index is widely used. In this work we studied and compared the performance of two different estimators of the concordance probability when a continuous predictor variable is categorised in a Cox proportional hazards regression model. In particular, we compared the c-index and the concordance probability estimator. We evaluated the empirical performance of both estimators through simulations. To categorise the predictor variable we propose a methodology which considers the maximal discrimination attained for the categorical variable. We applied this methodology to a cohort of patients with chronic obstructive pulmonary disease, in particular, we categorised the predictor variable forced expiratory volume in one second in percentage.